All Aboard Studios Route- Crewe To Birmingham New Street.

Ehi guys are the 350s formations correct for the 2019 timetable? Because last time, I was driving a service from Crewe to Euston and I was quite surprised by seeing that it used only one set. I don't know, it doesn't feel right.

 

All I can say is that the more I explore this route - the more impressed I am. Words just don't do it justice, but there is so much detail from start to end...the complexities of Crewe and Birmingham New Street (and Stafford, Wolverhampton), all the yards (Oxley Car is a real bonus too) - great countryside and the Black Country feels as it should do...even seeing traffic on the M5 at Sandwell and the M6 near Stafford....there is so much to look at...iron out the bugs for everyone, but this is a work of art !!

 

The little details stand out … bridge numbers, limited clearance signs, bits of fence lying on their side, the M6 lane closure sign, pedestrians on the canals… it’s been a pleasure exploring today! It’s obviously got a strong team behind it.

I do feel that someone was shooing them out the door before they were quite ready - e.g. why bother fully modelling Oxley and even including an ECS run to New St if it’s only going to start at Wolverhampton? Not enough time to model the signalling, perhaps?

If that’s true I hope that a) whoever is responsible honestly reflects on that counterproductive approach, and b) AAS gets a chance to finish it the way they originally wanted to!

I hope that Oxley doesn’t go the way of Toton…

 

I believe Adam or Kezz said to me on the Discord they are sorting that out as I said the exact thing about Oxley. It really is annoying as it breaks that immersion. But I hope they fix that as feels a big miss not doing ECS Moves in and out of Oxley.

 

Looking at some of the trackwork detail (the track clips most noticeably) it also looks as if some of it may be custom asset built? I noticed it particularly at Crewe, it looks very good.

 

But it also behaves like this when the signal is green.
When you depart a couple of seconds before the scheduled time. The ‘close doors’ message reappears when it’s the actual departure time. only stopping and reopening and closing the doors will then cancel the message.

 

I wish TSW could be inclusive of all consoles but the fact of the matter is the average modern pc is way more powerful than an 8th gen console and moderately more powerful than a 9th gen one. The PC version of TSW began having much more issues when DTG started having to develop multiple versions of timetables to include consoles. We’ve also seen timetables absolutely butchered (e.g. NY Trenton and WCMLS) with only a fraction of the services they should have because “well consoles wouldn’t be able to handle it, so all of you have to deal with a barebones timetable. We’re reaching the point where PC players are having to sacrifice what the game could be and deal with more rushed dev cycles in the effort to make basically three versions of the game so people rocking a 15 year old console can still keep up with a 1 year old pc.

 

I mean if you have an 8th gen console, or even a Series S, then I think you have to be realistic about the capabilities.

If you have a PS5 or Series X (like me!) then I think you are entitled to expect certain things still. Some of the games I have played this year on my X have been jaw dropping graphically and there is no reason why tsw can't still run perfectly well on them.

 

This exactly, I don’t understand certain members of the forum believing that the reason for unfinished timetables is due to Gen 8, when it clearly isn’t.

Considering the fact that certain routes have been released without Gen 8 included, timetables such as WCMLS aren’t barebones because of Consoles, they’re barebones because they are unfinished timetables.

It’s not as if the 2025 Class 805 WCMLS timetable being released by All Aboard Studios is just being released on PC.

Gen 8 is clearly not holding the game back if routes are able to be released on Gen 9 & PC without Gen 8 included.

 

It seems to me that the good comes with the bad. AAS are being praised, rightfully, for making a highly detailed, realistic recreation of the RL route with the full RL timetable: but all that detail, all that traffic, all those high-res textures and foliage and clutter, come at a performance cost. People, if you are going to have heavily packed content, you are going to get framerate drops and, in Unreal Engine, low LOD (blurry) texture, unless you are on a beast of a machine.

We are demanding ever more fidelity and detail, but want to have our cake and eat it without indigestion.

 

I would argue it’s the responsibility of the developer to actually remain within the confines of their game engine & the specs of their users.

Essentially killing the usability of your game for a massive percentage of users, in the pursuit of extra details seems self destructive & id be surprised if that’s a decision process that’s happened here. In fact from the various communications we’ve had from both AAS & DTG, it’s clear this outcome wasn’t intended.

The issue as always seems to be the testing is minimal & rushed, with little likely being done in response to any internal QA feedback.

AAS could have come up with best route concept we’ll ever see for TSW & on paper it’s right up there. Unfortunately when this release is the reality, that’s all meaningless. AAS will now be in a position where at least their next couple of products will rightfully be questioned on the back of the poor condition of this release, all because no one seemingly checked what was actually being released.

You say fidelity & detail are the demands of the customer, but I’d argue there’s been a much, much bigger demand for better QA, less repeat mistakes & actually shipping the correct build.

 

You are rather missing the point. It wasn't DTG who packed this release to the gills.

 

Where did I say it was?

 

You are blaming DTG for the state of the release. But it seems to me that, other than shipping the wrong console builds, none of the other issues were DTG's doing. It was AAS who built a DLC too heavy for many machines to handle.

 

But UK signalling would also normally be set before the departure time. In the current timetable of this route, there is a load passengers until departure time, which then triggers a second load passengers task which causes the signal and route to be set - this is what the person you quoted is referring to.

Hopefully the timetable can be updated so that the route and signals are set for the "proper" departure time of say 10:45, rather than 10:45 just starting a new load passengers task to set the signal.

 

With hindsight and they probably would never admit to it, Birmingham to Crewe was far too ambitious for a first route project. Whether this was political to get one over on ATS or a simple case of aspiration exceeding capability again we will never know. But while delivering what cannot be denied is a nice looking route, customers have been caught in this developer crossfire as the key elements that make the route work, fall short. Reading between the lines of what was said yesterday, I wonder if the AI SPAD issue can be fixed at all and while work is underway to fix the timetable, if the knowledge as to how it should be set up in the first place to avoid missing arrival times, double despatch and way off predicted running times wasn’t there, what has changed? And if it has been a learning experience that shouldn’t have been at the expense of customers paying £30.

Moral of the story is, start short and relatively simple. I mean Skyhook’s MML is probably an example of a more modest prototype and even they botched it slightly on the first pass. Picking a route that features two of the largest and most complex railway locations in the UK was a recipe for disaster, if you’re not entirely sure what you’re doing.

Anyway going to have a little side bet, that as early as the Christmas Steam sale we will be seeing this route getting a 20% off discount. At which point, if the promised fixes are delivered and everything works as it should, might be worth a flutter at £24.

 

There's a lot more to testing than QA. QA should be the last line of defence, picking up on silly problems, bashing things in unexpected ways to see how it's handled, etc. If the code is arriving at QA in the state that this has, then it's far too late to realistically do anything to fix it except scrap the release and go back to the drawing board. The trouble is, the same problems will keep happening over and over again unless the fundamental systems are in place to ensure that these problems are caught early and often. From my decades of experience, it feels like the fundamentals of quality and testing are not in place. This doesn't come as a complete surprise to me because having worked with people who have come from the gaming industry, things are often lashed up and chucked out of the door and then the codebase is largely forgotten about as they essentially reimplement or copy over large swathes of it for the next iteration, doing enough to make it work this time around. This is no slight on the individual developers, artists, QAs, support people, (and yes) managers, etc. involved because, as individuals, they are often super-talented, conscientious, hard-working people, but if quality isn't at the heart of the culture, they are howling at the moon, and I know several people who have left the industry for entirely that reason. TSW is different from a lot of other games. It's what my company calls a 'heritage' product, with years and years of history, compromises, technical debt, etc. behind it but it has to keep trudging forwards somehow. This all needs careful management to establish a quality culture early on because it gets increasingly difficult to retrofit quality at the centre of a codebase as the years go by. This is all conjecture though. I don't know how DTG manage their codebases, all I can do is comment on what I can see happening and try to align that with my experiences of software development, both good and not so good. And, whatever's going on, I have a lot of sympathy for the individuals trying their best to make things work for their customers.

 

Makes you wonder why they didn't try the Class 805 pack for WCMLS first as that would have given experience of both a new a train and building a highly complex timetable before tackling the bigger challenge of complex route assets on top. I suppose there's a lot of hindsight to be had on all this.

 

I think the concept of them starting smaller entirely misses the point of what's gone wrong here.

 

Agree with you there Vern, however I'd be amazed if we see a 20% discount in just 2 months time!

I'm holding off purchasing this now anyway until the New Year, let all, if not most of the problems get ironed out, should be an excellent route in a few months

 

I'll repeat myself, but AI SPAD is practically a problem of 2 stopping places versus signal position on this map.......I don't believe that the stopping position can't be moved a few meters......then that would have to be a problem always and everywhere and I don't feel like it's happened anywhere like this before or was it a miracle that it hasn't happened elsewhere? I've done 1800 miles on WCML South, I don't know about having AI SPAD

 

"Slightly"?!? The sound and fury at this route is nothing compared to the absolute (bleep)storm there was around MML when it first came out.

(EDIT: it seems even the letter s followed by 3 stars falls foul of the profanity filter!)

 

I don’t believe I specifically laid blame to anyone in my post, other than the ‘developers’. I’ll certainly clarify that though & say they’re both to blame.

In terms of what’s actually been built, I don’t think there’s anything that’s been done here that pushes the route beyond anything else in TSW. We’ve had busier timetables, bigger stations, more layers & denser scenery - in other releases.

In terms of what’s actually released, AAS have said it was DTG’s responsibility when it came to QA & that’s where the brakes should have been put on. If what we’ve got is what DTG were supplied with to test, the whole thing should have been kicked back to AAS to sort out.

I agree with Vern in saying there was likely too much here for AAS to chew for a first release, but then if DTG just sign off on it & say it’s good for release, then that’s all AAS can go off. Personally I think it’s daft that no one at any point made AAS aware of the issues (so they’ve said) & that AAS didn’t do their due diligence when it came to verifying this, but clearly they had misplaced trust in DTGs QA.

We could go back and forth with who’s to blame for each specific thing here, but at the end of the day it’s DTG who went ahead & published the route after failing to feedback the problems or told AAS there weren’t any. Though I think individual blame is irrelevant, both DTG & AAS had responsibilities they’ve failed to hold up.

If the primary issue at play here is what you determined it to be, a case of too much fidelity & detail, then there should have been something relayed to customers about the severe consequences of that in this particular DLC, instead we were told not to expect any issues & it was advertised the same as any other DLC.

I also don’t see how your reasoning leads to things such as AI’s causing player SPADs, a borked timetable, or any of the other numerous technical issues in play. It’s just another case of not enough time & insufficient testing.

 

This was a point I’d made at some point prior to release, I think it’s certainly a valid one when a team that self admittedly, doesn’t have enough staff to even test the product, is attempting to make a rather meaty route, along with essentially rebuilding one train & developing a timetable + new train for another route. That just sounds like far too much too quickly. I think AAS are going to have some rather poor feedback when the 805 announcement comes around, I suspect that most issues with this route will still be present when they start pushing that.

That SPAD issue is certainly one of the more severe bugs that TSW has ever had, you’re basically playing russian roulette with the game when picking a service & the fact that this issue is something that apparently can’t be detected during testing (though seemingly detectable issues don’t get resolved anyway) is quite concerning.

I would have made a point of some potential prior experience of AAS’ devs in defence of the scope they’ve gone for, but I agree there was obvious lack of experience & that’s very obvious with the timetable. When you look back at what other third parties have done & issues they encountered, with small scale routes & livery packs etc, I do find it a little egotistical that AAS decided a 50+ mile route between 2 large, busy termini, with a huge timetable & rebuilds of the 350 family would be their first project. All whilst building another loco DLC & timetable (and who knows what else they’re working on). The comments they made about ATS’ development cycle back when MAC was in the workings has certainly left them in an ‘egg on face’ situation too.

That being said, you have to wonder where DTG’s 3PP was in all of this. They made a real song & dance out of all the guidance & support they offer during the Mildmay stream, but here it seems that they essentially let AAS walk straight into a bear trap.

As I said above, on paper this DLC is certainly something great, but whether it will ever get brought upto that standard in practice is something I’m not sure on. In any case there certainly needs to be reflection on AAS’ side when it comes to time & testing needed.

 

Sorry for the late response but yes, it came through the next day and fixed the issue so thank you

 

In my view, the problem is the result of poor or nonexistent testing, mostly on the part of DTG, which has the resources to thoroughly check for issues before release.

I don't blame AAS at all for being ambitious and wanting to make the best route they could. In fact I applaud them for making a brave attempt to advance the game and introduce more sophistication, things in which DTG should be leading the way. My only criticism is that it is not ambitious enough in that it should have included at least one new train set.

Upon receiving the route, it was up to DTG to test it and return it for correction if necessary, not rush it out with the issues we're seeing now. They should have paid particular attention to its performance on console. That's what publishers do, whether it's a novel, history work or video game. Publish and be damned might work for a political memoir, but not for a simulation. Any one of us could have played the route for a couple of hours and easily found most, if not all, of the problems. Which is basically what happened. Didn't anyone at DTG do this? On an average PC? On a gen9 console?

It's clear to me that the whole testing regimen at DTG is like Swiss Cheese, if it exists in any real way at all.

 

Well, there are some reports that this route was ready for release more than 3 months ago for TSW5. If, and I stress if, that is the case, it's even more mystifying that it has so many probkems. What was happening during this period?